Hydroponic culture is a plant cultivation technique in which plants are grown in a nutrient solution without soil. It has the potential for high crop productivity in a small area. However, if plant pathogens enter the solution, they can spread rapidly throughout the hydroponic culture facility and cause catastrophic damage. For this reason, disinfection of the solution is essential, although direct administration of pesticides into the solution has been prohibited at law (
Ministry of Agriculture, Forestry and Fisheries, 2010). Therefore, it is desired to establish safe and effective alternative disinfection methods, and the various disinfection treatments have been investigated, including UV light, heat, the titanium dioxide photocatalytic reaction, and ozone (O
3) (
Bando et al., 2008;
Dannehl et al., 2016;
Ehret et al., 2001;
Igura et al., 2004;
Koohakan et al., 2003;
Ohtani et al., 2000;
Runia, 1995). However, these treatments still are not put into any practical use in terms of the efficiency, treatment time, and running cost. Particularly, O
3 gas is effective disinfectant due to its strong oxidation power, although it is difficult to use in hydroponic cultures because of its extremely low solubility in water (0.105 g 100 ml
−1 (0°C)).
Recently, tiny bubbles less than 50 μm in diameter, called microbubble (MB), have been studied and used in many fields. They rise more slowly in water than millibubbles (MMB), which have diameters in the mm to cm range (
Takahashi et al., 2003;
Takahashi, 2005), and possess additional properties such as the interface charge, long stagnation, slow buoyancy, the shrinkage and the generation of free radicals by their collapsing other than dissolving power (
Li and Tsuge, 2006;
Li et al., 2009;
Zheng et al., 2015). Previously, we focused on long retention time in water and the dissolving power of MB in addition to the strong oxidative power of O
3, and found that O
3MB were more effective than O
3MMB for the disinfection of
Fusarium oxysporum f. sp.
melonis spore and
Pectobacterium carotovorum subsp.
carotovorum in nutrient solution for hydroponic culture (
Kobayashi et al., 2011a,
2011b,
2011c,
2012). Furthermore, effective disinfection by O
3MB has been reported by other researchers (
Chuajedton et al., 2015;
Inatsu et al., 2011), although it is still not clear about the exact mechanism of disinfection by O
3MB. In this study, the morphological change of
F. oxysporum f. sp.
melonis spores by O
3MB and O
3MMB was therefore observed by using scanning and transmission electron microscopies (SEM and TEM), and the difference was discussed.
F. oxysporum f. sp.
melonis NBRC6385 suspension (approximately 1.0 × 10
7 spores ml
−1) was prepared in a manner similar to a previous report (
Kobayashi et al., 2011a). For each experiment, 15 l of tap water were collected in a plastic cylindrical container (28 cm dia. × 48 cm height) and kept for 24 h at room temperature to remove chlorine. Water quality test paper (Nissan aquacheck 3; Nissan Chemical Industries, Ltd., Tokyo, Japan) was used to confirm that no residual chlorine remained. O
3 was generated by using an O
3 generator (ED-OG-A10, Ecodesign Co. Ltd., Saitama, Japan) at a flow rate of 2.5 l min
−1. O
3MB and O
3MMB were generated by using a decompression type MB generator (20NEDO4S, Shigen-Kaihatsu Co., Ltd., Kanagawa, Japan) and a commercial air pump, respectively. The concentration of dissolved O
3 in both O
3MB and O
3MMB waters was set to 1.5 ppm at 15°C. The pH of O
3MB and O
3MMB waters was the same at 6.8 and was not changed before and after O
3MB and O
3MMB generation.
F. oxysporum f. sp.
melonis spores were added to 100 ml of the O
3MB and O
3MMB waters with final concentrations of 1.0 × 10
3-1.0 × 10
4 cfu ml
−1. Aliquots of the treated waters were collected after 0, 15, 30, 45, 60 and 120 s. Aliquots of 0.1 ml of the collected waters were plated on potato dextrose agar (Difco, Becton Dickinson, Flanklin Lakes, NJ, USA) plates, and the plates were incubated at 30°C for 48 h. After incubation, the numbers of surviving spores were measured by counting the colonies formed on the plates. The detection limit was 10 cfu ml
−1. Each experiment was performed in duplicate.
F. oxysporum f. sp. melonis spores were collected from 5 ml of the suspension by filtration with a cartridge filter (Anotop 10, GE Healthcare UK Ltd., Buckinghamshire, UK). The sample on the cartridge filter was pre-fixed with 2.5% glutalaldehyde solution diluted with a phosphate buffer solution (PBS, pH 7.0). A filter ejected by decomposing the cartridge filter was washed with PBS (pH 7.0), post-fixed with 2% OsO4 solution for 1 h, and then serially dehydrated for 20 min each in 50%, 70%, 80%, 90%, 95%, 99.5% and dehydrated ethanol. SEM observations were performed as follows: The dehydrated sample was immersed in a mixture of t-butyl alcohol and dehydrated ethanol (1:1) for about 10 min, transferred to 100% t-butyl alcohol, freeze-dried with a freeze drier (ES-2030, Hitachi High Technologies Co., Tokyo, Japan), and OsO4-coated with a OsO4 coater (HPC-1SW, Vacuum device Inc., Mito, Japan) (the thickness of the coating was adjusted to 3 nm). Then the sample was observed with an SEM (JSM-6700F, JEOL Ltd., Akishima, Japan) operated at 3.0 kV. Nine F. oxysporum f. sp. melonis spores in the SEM photographs were selected at random and the widths was measured with a scale. Significant differences were evaluated by the ANOVA and Fisher’s LSD using the Ekuseru-Toukei 2012 for Window statistical software (Social Survey Research Information Co., Ltd., Tokyo, Japan) (P < 0.05). TEM observations were performed as follows: The dehydrated samples were serially immersed for 2 h each in 1:1, 2:1, and 3:1 mixtures of Quetol-651 (Cosmo Bio Co., Ltd., Tokyo, Japan) and ethylene glycol diglycidyl ether, and then embedded in 100% Quetol-651 at 60°C. Ultra-thin sections (thickness 70-100 nm) were made from the embedded samples with an ultramicrotome (ULTRA CUT UCT, Leica Microsystems, Wetzlar, Germany). The ultra-thin sections were doubly electron-strained using 4% uranyl acetate for 12 min and lead nitrate for 5 min, and then observed with a TEM (JEM-2010, JEOL Ltd.) operated at 140 kV.
The survival rates of
F. oxysporum f. sp.
melonis spores in water treated with O
3MB and O
3MMB are shown in
Fig. 1. The disinfectant efficiency of O
3MB on
F. oxysporum f. sp.
melonis spores was greater than that of O
3MMB, because the numbers of surviving spores after treatment with O
3MB and O
3MMB reached the detection limit at 45 s and 60 s, respectively. The result agreed with our previous study (
Kobayashi et al., 2011a). Amount of hydroxyl radicals generated from O
3 is not enough to have a disinfectant effect (
Cho et al., 2003). However, O
3MB may generate more hydroxyl radicals than O
3MMB, because the oxidation-reduction potential and iodine liberation is higher with O
3MB than with O
3MMB (
Chuajedton et al., 2015). Furthermore, the use of MB enhances the formation of hydroxyl radicals due to more rapid O
3 decomposition (
Tsuge et al., 2009), and the hydroxyl radicals generated from O
3MB accelerate the oxidative power (
Chu et al., 2008). The high oxidative power of O
3MB contributes to the oxidative power of O
3, the reactivity of the hydroxyl radicals, the substantivity, ζ surface potential, mass-transfer coefficient, and use efficiency (
Zheng et al., 2015). Therefore, the greater disinfectant efficiency of O
3MB than O
3MMB is likely be due to these synergistic effects.
The SEM images of
F. oxysporum f. sp.
melonis spores treated with O
3MB and O
3MMB are shown in
Fig. 2. The spores treated with O
3MMB for 30 s showed no obvious surface injury, although spores treated with O
3MMB for 180 s were deformed. On the other hand, spores treated with O
3MB showed obvious surface injury after 30 s and the spores were completely destroyed after 180 s. The widths of
F. oxysporum f. sp.
melonis spores treated with O
3MMB for 30 s were lower than those of non-treated spores, and then the spores swelled after 180 s (
Fig. 3). However, the widths of spores treated with O
3MB for 30 s were the same as those of non-treated spores and diminished in size by 180 s. Furthermore, the TEM images indicated the appearance of liquid foam in the mid-regions of the
F. oxysporum f. sp.
melonis spores treated with O
3MMB for 180 s (
Fig. 4). Then, the spores had swelled by O
3MMB for 180 s, as water entered the cells through the damaged cell wall.
Cho et al. (2010) reported that disinfection of bacterial cells by O
3 was due to injury of the cell wall.
Zhang et al. (2011) showed that disinfection of
Pseudomonas aeruginosa by O
3 was due to increase in cell membrane permeability and coagulation of the intracellular substrate.
Thanomsub et al. (2002) concluded that disinfection by O
3 was caused due to the destruction of the cell wall and leakage of the intracellular substrate, followed by cell lysis. Therefore, it is possible that the
F. oxysporum f. sp.
melonis spores treated with O
3MMB for 30 s initially shrank due to leakage of the intracellular substrate through the damaged cell wall. On the other hand, the TEM images of the
F. oxysporum f. sp.
melonis spores treated with O
3MB for 180 s showed the wavelike deformation of cell membrane and appeared to have a space between the cell membrane and/or wall and the cytoplasm.
Diao et al. (2004) confirmed that hydroxyl radicals generated by the Fenton reaction induced the injury of
E. coli cell membranes greater than O
3 by the observation with SEM. Therefore, it appears that hydroxyl radicals generated from the O
3MB induces the wavy injury of cell membrane of
F. oxysporum f. sp.
melonis spores. Furthermore, it was considered that coagulation of the substrate within the spores and leakage of the substrate through the damaged cell membrane were induced by the higher amounts of hydroxyl radicals generated from the tiny O
3MB penetrated into the spores. The result may lead to cell death due to the leakage and/or coagulation of the intracellular substrate, followed by the lysis of the spore.
These results show that disinfection efficiency of O3MB on F. oxysporum f. sp. melonis spores is higher than that of O3MMB and may be due to the action of MB in combination with the high oxidative power of the O3. Therefore, it is considered that disinfection of F. oxysporum f. sp. melonis spores by O3MB causes the leakage and/or coagulation of intracellular components associated with damage to the cell membrane and/or cell wall, and subsequently leads to lysis of the spore.